### TABLE I Performance comparison of image quality assessment models. CC: correlation coefficient; MAE: mean absolute error; RMS: root mean squared error; OR: outlier ratio; WMAE: weighted mean absolute error; WRMS: weighted root mean squared error; SROCC: Spearman rank-order correlation coefficient

2004

Cited by 97

### TABLE I Performance comparison of image quality assessment models. CC: correlation coefficient; MAE: mean absolute error; RMS: root mean squared error; OR: outlier ratio; WMAE: weighted mean absolute error; WRMS: weighted root mean squared error; SROCC: Spearman rank-order correlation coefficient

2004

Cited by 33

### Table 2: Resistance to Outliers. Classification error. Outlier

2000

"... In PAGE 5: ... Again, cross val- idation was used to select a109 and a35 . Table2 presents the generalization performance with each row being an average over ten independent runs. The table indicates that Parzen Windows hyperplanes with Gaussian kernels tend to be more resistance to outliers than the other linear methods.... ..."

Cited by 13

### Table 2: Resistance to Outliers. Classification error. Outlier

2000

"... In PAGE 5: ... Again, cross val- idation was used to select and C. Table2 presents the generalization performance with each row being an average over ten independent runs. The table indicates that Parzen Windows hyperplanes with Gaussian kernels tend to be more resistance to outliers than the other linear methods.... ..."

Cited by 13

### Table 2: Resistance to Outliers. Classification error. Outlier

2000

"... In PAGE 5: ... Again, cross val- idation was used to select and C. Table2 presents the generalization performance with each row being an average over ten independent runs. The table indicates that Parzen Windows hyperplanes with Gaussian kernels tend to be more resistance to outliers than the other linear methods.... ..."

Cited by 13

### Table 1: Estimated translation vectors. ttrue is the true translation. No stands for the number of arti cial outliers added to data. Estimator identi es the estimator used to compute the motion. (LSQ = Least Squares Estimator, R(oi) = robust estimator using oi measure to sort the correspondences). terror = ttrue ?testimated. qE labels the quality of selected solution and n gives the number of best ranked

"... In PAGE 13: ... The errors in rotation were less than one degree in Euler angles representing the rotation. Estimated translations are listed in Table1 . To compare estimated translations with the true ones, the length of estimated t is set to its true the length.... ..."

### Table 2: Estimated translation vectors and angle. ttrue is the true translation and r is true Euler angle in degrees. No stands for the number of arti cial outliers added to data. Estimator identi es the estimator used to compute the motion. (LSQ = Least Squares Estimator, R(oi) = robust estimator using oi measure to sort the correspondences). terror = ttrue ? testimated. e = r ? estimated. qE labels the quality of selected solution and n gives the number of best ranked correspondences used to compute the solution.

"... In PAGE 28: ...rom the scene in the distance of about 1.8 m when pure translation is cosidered. In the case of general motion much higher errprs were encountered. However, it is apparent from Table2 , that these errors are not due to miscorrespondences since even wors results are obtained when LSQ estimatir is used. Therefore, we believe that manual extraction of correspondences was a ected by large error.... ..."

### Table 4. IP/Port frequency distribution

2004

"... In PAGE 10: ...vide higher quality outlier scores. To illustrate this consider the following probability distribution ( Table4 ), where blank represents no occurrences. If we observe a packet P1 on IP2/Port3 and another packet P2 on IP4/Port1, SPADE will assign equal anom- aly scores to both of the packets.... ..."

Cited by 1

### Table 1. Outliers Ranking

2006

"... In PAGE 7: ...2 Due to the limitation of space, we only show two instances. Table1 lists the top 6 outliers based on the sample dataset in Figure 5, by both LOF and INFLO measures. The most outstanding outliers can be recognized by... ..."

Cited by 2

### Table 1. Performance comparison of image quality assessment models on LIVE JPEG/JPEG2000 database [13]. SS-SSIM: single-scale SSIM; MS-SSIM: multi-scale SSIM; CC: non-linear regression correlation coefficient; ROCC: Spearman rank-order correlation coefficient; MAE: mean absolute error; RMS: root mean squared error; OR: outlier ratio.

2003

Cited by 13